Multi-objective optimisation of agricultural product distribution paths using GA-BP neural network for green logistics

In the process of agricultural product distribution, there are problems such as low distribution efficiency, high cost, and high carbon emissions. This paper combines historical enterprise logistics data and real-time traffic data to establish a genetic algorithm-back propagation (GA-BP) neural network model. It optimises the distribution route of for agricultural products based on multiple targets of: time, cost, and carbon emissions. Compared with empirical methods, greedy algorithms and nearest neighbour algorithms, the distribution efficiency is improved by 22%, logistics cost optimisation rate up to 55%, and carbon emissions are significantly reduced, which further verifies the effectiveness of green logistics.